A Canadian graduate student is mapping brainpower to computer image searches. It’s expected to be a novel way to empower search engines for visual images … and should be available this summer.
According to the University of Ottawa graduate student, Kris Woodbeck, Canadian government officials at Technology Transfer and Business Enterprise (TTBE) will help him secure a patent on his innovative computer search approach.
The invention earned Woodbeck the university’s Innovator of the Year award last week. His search engine copies the brain’s approaches to process visual information. How does it work?
“The brain is very parallel. There’s lots of things going on at once,” he said. “Graphics processors are also very parallel, so it’s a case of almost mapping the brain onto graphics processors, getting them to process visual information more effectively.”
Currently, Woodbeck is testing his search engine technology on academic data sets that include between 60,000 and 100,000 images. He hopes to classify these according to certain labels.
He’ll then use Web crawlers to index images so that his search engine will have a critical mass of content to launch. Can you use more efficient image related Web searches where you work?
Comment Preview